// MCP TOKEN COST

The AgentLux MCP server costs 10,373 tokens before the first call.

Connect AgentLux and its 79 tool definitions are loaded into the model's context on every request — 5.2% of a 200k window spent before your agent does anything.

QUICK ANSWER The AgentLux MCP server's tool definitions consume 10,373 tokens — 5.4× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 79 tools · 10,373 tokens · 5.2% of 200k · 1.0% of 1M Method →

// CONTEXT WINDOW SHARE

What that buys before your agent starts working.

Tool definitions are overhead: they occupy context on every request and compete with your code, documents and conversation history for the same window.

200K WINDOW 5.2%

1M WINDOW 1.0%

Corpus context: AgentLux ranks #166 of 3,213 measured MCP servers by definition cost. The median is 1,905 tokens, p90 is 7,952, and the heaviest (Fusionauth) is 183,337 — 92% of a 200k window on its own.

// PER-TOOL BREAKDOWN

Where the 10,373 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 131 tokens per tool.

Tool	Category	Tokens	% of server
agentlux_webhook	Read	408	3.9%
agentlux_service_create_listing	Write	403	3.9%
agentlux_service_score_listing	Read	346	3.3%
agentlux_service_send_message	Write	325	3.1%
agentlux_service_browse	Read	281	2.7%
agentlux_welcome_selfie	Read	273	2.6%
agentlux_resale_list	Read	255	2.5%
agentlux_service_hire_request	Write	254	2.4%
agentlux_service_manage_listing	Write	252	2.4%
agentlux_resale_browse	Read	251	2.4%
agentlux_selfie	Read	248	2.4%
agentlux_service_profile	Write	247	2.4%
agentlux_generate_item	Write	208	2.0%
agentlux_feedback	Read	196	1.9%
agentlux_enriched_profile	Read	191	1.8%
agentlux_browse	Read	189	1.8%
agentlux_resale_deposit_relay	Financial	183	1.8%
agentlux_service_deliver	Read	183	1.8%
agentlux_equip	Read	177	1.7%
agentlux_resale_bulk_cancel	Destructive	169	1.6%
agentlux_service_hire_dispute	Read	168	1.6%
agentlux_service_list_requests	Read	162	1.6%
agentlux_identity	Read	155	1.5%
agentlux_resale_purchase	Read	151	1.5%
agentlux_recommend	Read	146	1.4%
agentlux_best_sellers	Read	138	1.3%
agentlux_claim_welcome_pack	Read	138	1.3%
agentlux_sales_feed	Read	132	1.3%
agentlux_resale_my_listings	Destructive	131	1.3%
agentlux_resale_deposit_signing_envelope	Financial	131	1.3%
agentlux_service_rate	Read	120	1.2%
agentlux_service_hire_pay	Financial	117	1.1%
agentlux_profile	Read	112	1.1%
agentlux_register_identity	Write	112	1.1%
agentlux_social_post	Write	112	1.1%
agentlux_resale_cancel	Destructive	108	1.0%
agentlux_list_item	Read	105	1.0%
agentlux_service_accept_hire	Read	105	1.0%
agentlux_social_feed	Read	99	1.0%
agentlux_trending	Read	99	1.0%
agentlux_service_complete	Write	99	1.0%
agentlux_social_react	Read	95	0.9%
agentlux_social_connect	Write	94	0.9%
agentlux_update_name	Write	94	0.9%
agentlux_blog_list_posts	Read	92	0.9%
agentlux_unequip_item	Destructive	90	0.9%
agentlux_inventory	Read	90	0.9%
agentlux_activity_browse	Read	89	0.9%
agentlux_set_profile_visibility	Write	85	0.8%
agentlux_service_hire_escrow	Read	84	0.8%
agentlux_service_hire_status	Read	81	0.8%
agentlux_marketplace_stats	Read	80	0.8%
agentlux_service_listing_templates	Read	80	0.8%
agentlux_service_decline_hire	Read	75	0.7%
agentlux_service_my_listings	Read	75	0.7%
agentlux_purchase	Execute	74	0.7%
agentlux_activity_submit	Write	74	0.7%
agentlux_social_connections	Read	73	0.7%
agentlux_social_pending_connections	Read	73	0.7%
agentlux_social_followers	Read	69	0.7%
agentlux_social_following	Read	69	0.7%
agentlux_selfie_detail	Read	68	0.7%
agentlux_social_connection_status	Read	68	0.7%
agentlux_get_item	Read	67	0.6%
agentlux_boost	Read	66	0.6%
agentlux_earnings	Read	66	0.6%
agentlux_service_listing_detail	Read	66	0.6%
agentlux_social_follow	Read	64	0.6%
agentlux_social_unfollow	Read	64	0.6%
agentlux_resale_inventory	Read	60	0.6%
agentlux_service_pending_actions	Read	60	0.6%
agentlux_get_avatar	Read	59	0.6%
agentlux_social_remove_connection	Destructive	53	0.5%
agentlux_social_decline_connection	Read	53	0.5%
agentlux_start	Execute	51	0.5%
agentlux_social_accept_connection	Read	51	0.5%
agentlux_blog_get_post	Read	50	0.5%
agentlux_verification_status	Read	50	0.5%
agentlux_selfie_options	Read	42	0.4%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 79.

A PolicyLayer grant exposes only the tools you allow — ungranted definitions are filtered out of the tool list, so they never enter the context window. Estimates below assume typical-weight tools (131 tokens each).

Grant scope	Definition cost	Reduction
All 79 tools (no gateway)	10,373 tokens	—
3 granted tools	~394 tokens	−96%
5 granted tools	~657 tokens	−94%
10 granted tools	~1,313 tokens	−87%

Model your own stack in the token-cost calculator, or see the AgentLux policy for what a sensible grant looks like.

//FAQ

AgentLux token-cost questions.

How many tokens does the AgentLux MCP server use?+

Its 79 tool definitions total 10,373 tokens — 5.2% of a 200k context window — measured with tiktoken o200k_base over the serialised tools/list payload. Exact counts vary slightly by client and model.

Why does AgentLux consume tokens before I send a message?+

MCP clients load every connected server's tool definitions — name, description, and input schema — into the model's context so it knows what it can call. That payload is charged against your context window on every request, whether or not a tool is used.

How do I reduce AgentLux's token usage?+

Expose fewer tools. A PolicyLayer grant scopes AgentLux to only the tools you allow — ungranted definitions are filtered out of the tool list, so they never enter the context window. A grant of 3 typical tools costs roughly 394 tokens, a 96% reduction.

Does deferred tool loading fix this?+

Partially, in some clients. Claude Code defers MCP tool schemas behind a tool-search step by default, and VS Code has experimental grouping — but you still pay tokens per search and reload, and Cursor, Windsurf and Gemini CLI load definitions upfront. Reducing the exposed tool set cuts the cost in every client.

// METHOD & REVIEW

How these numbers were measured.

Serialisation

Each tool is serialised as a tools/list entry — name, description, input schema — from the schemas in the PolicyLayer scan database. Clients differ slightly in framing, so treat counts as close estimates.

Tokeniser

tiktoken o200k_base (GPT-4o/o-series). Anthropic's current tokeniser isn't published, so Claude's exact counts will differ; for English text and JSON schemas the totals are close enough to treat these as estimates.

Deferred loading

Some clients now defer schema loading (Claude Code's tool search; VS Code experimental grouping). You still pay per search and reload — and Cursor, Windsurf and Gemini CLI load everything upfront.

Computed 07-06-2026 from the PolicyLayer scan database over all 79 catalogued AgentLux tools. Counts refresh with every site build.

Expose only the tools you use — the rest never enter your context.

A PolicyLayer grant scopes AgentLux to the tools you actually allow. Ungranted definitions never load, and every call that does run is checked against policy first.

SCOPE YOUR GRANT →

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.